insp

> 18  20  22  24  Sorry. I set --init-word== 'cat' for this, not --init-word charactor. The model.ckpt is sd-model-v4, because I think output .pt file can...

> Oh, that's what I just learned this. thanks! By the way,I see your work in [##](https://github.com/rinongal/textual_inversion/issues/35) is very good. Do you have consistent documents or notes to train a...

@genekogan I add arg "--no-test" to make ignore the err. @1blackbar [Hugging](https://huggingface.co/sd-concepts-library/tomcat) that' origin image.I do trained it simply in google colab. and the results were equally unsatisfactory.

After I try 6100 steps, I think it can continue 3000 times to see the effect. How to restart training from previous end point?

> add `-embedding_manager_ckpt logs/images_yourDate_yourName/checkpoints/embeddings.pt` Thanks! nice args!

> Hi, I was wondering if you had an update on if you were able to get better results? I've changed using real person training recently. But I found only...

> @inspire-boy please, try to: > > ``` > pip install --upgrade pip > pip install openvino > ``` Its like everything OK, BUt in my Win10: PS D:\py\stable_diffusion.openvino> python...

my cpu is Intel core(TM) i7-6700 @ 3.4GHz

> Seeing the exception my first thought was maybe openvino doesn't support your CPU, so I checked the requirements and it says 6th to 12th generation is supported. Yours is...

This is very good! by the way, Is there a synchronous example? just one request wait for one img response? @comfyanonymous